sweezin

followers

following

stars

sweezin's repositories

PI-DECODER

precise iris segmentation

Language:PythonApache-2.010 30

mtcnn

基于caffe的mtcnn训练实现，可以训练一个自己的有效的目标检测算法，非常容易非常简单，并且有配套的纯c++版本的mtcnn-light

Language:C++1 20

68-Retinaface-Pytorch-version

Working on 68/96 landmarks detection with RetinaFace

Language:Python000

awesome-c-cn

C 资源大全中文版，包括了：构建系统、编译器、数据库、加密、初中高的教程/指南、书籍、库等。

010

frvt

Repository for the Face Recognition Vendor Test (FRVT)

Language:C++NOASSERTION000

hello_tf_c_api

Neural Network TensorFlow C API

Language:C++MIT000

hiddenlayer

Neural network graphs and training metrics for PyTorch, Tensorflow, and Keras.

Language:PythonMIT010

HTTPRequest

Single-header C++ HTTP request class

Language:C++Unlicense000

HyperDL-Tutorial

深度学习教程整理 | 干货

000

libfacedetection

An open source library for face detection in images. The face detection speed can reach 1000FPS.

Language:C++NOASSERTION000

MNN

MNN is a lightweight deep neural network inference engine.

Language:C++010

model_deployment

sample model deployment code

000

nnie

重构海思sample中的NNIE模块

Language:C010

PFLD

Implementation of PFLD A Practical Facial Landmark Detector , reference to https://arxiv.org/pdf/1902.10859.pdf

Language:Python010

PFLD-pytorch

PFLD pytorch Implementation

Language:Python010

srmd-ncnn-vulkan

SRMD super resolution implemented with ncnn library

MIT000

tensorboardX

tensorboard for pytorch (and chainer, mxnet, numpy, ...)

MIT000

TNN

TNN：移动端高性能、轻量级推断框架，同时拥有跨平台、高性能、模型压缩、代码裁剪等众多突出优势。TNN框架在原有Rapidnet、ncnn框架的基础上进一步加强了移动端设备的支持以及性能优化，同时也借鉴了业界主流开源框架高性能和良好拓展性的优点。目前TNN已经在手Q、微视、P图等应用中落地，欢迎大家参与协同共建，促进TNN推断框架进一步完善。

Language:C++NOASSERTION010

YOLO-World

Real-Time Open-Vocabulary Object Detection

GPL-3.0000